only connect: how everything is linked to everything else

My teenage son keeps reminding me that everything is subject to the laws of physics. That idea is often disputed, but since there’s general agreement that autistic characteristics are caused by something in the physical world, I want to have a go at explaining how the laws of physics tie together everything in the physical world – including autism. How we believe everything is tied together affects how we categorise or classify things – including autistic behaviour.

The way human knowledge is organised by educational institutions, libraries and encyclopedias, you could be forgiven for thinking that the laws of physics might apply to the behaviour of solids, liquids and gases, but aren’t relevant to things like art, literature or human emotion. People don’t behave like subatomic particles, so thinking of them in terms of subatomic particles is often dismissed as reductionist, because it ignores the complexity and subtlety of thought and behaviour – there must be something qualitatively different about living organisms, especially humans beings.

Levels of complexity

The reason my son says everything is subject to the laws of physics is because subatomic particles (I’ll stick with ‘particles’ for simplicity) behave in consistent ways that we describe as ‘laws’. Everything in the physical world consists of subatomic particles in different configurations of varying complexity. Subatomic particles interact to form atoms. Different configurations result in different kinds of atoms – chemical elements. Atoms form molecules, ranging from simple chemical compounds to highly complex organic molecules. Organic molecules make up living organisms, and some of those organisms can paint, write and express their emotions.

Each different configuration of particles opens up some possibilities for change but closes down others. If a set of particles has formed an atom of gold, for example, its configuration is very stable, so there’s not much likelihood of it changing. That’s why we can dig gold out of the ground. An atom of sodium on the other hand, is in an unstable configuration, so there’s a lot of potential for change. Sodium readily combines with other elements to form salts – sodium chloride, carbonate and bicarbonate etc. Which is why don’t find chunks of sodium lying around underground – any pure sodium would have combined with other elements.

molecular structure of sodium chloride

Opportunities that a specific configuration opens up are called affordances, and opportunities that those configurations close down are called constraints. At different levels of complexity, different affordances and constraints come into play. Subatomic particles can form hydrogen, release vast amounts of energy and hold the universe together. Sodium atoms can’t do those things, but they can form chemical compounds, which single subatomic particles can’t. Human beings can’t transport oxygen between cells (not without a lot of equipment anyway); haemoglobin can, but it can’t paint a picture or write a book.

Because everything in the physical world is connected in this way, anything – from a physical object to the most sophisticated human activity – can be construed any level of complexity you care to choose. A bit like zooming up and down on Google Earth. Writing a novel, for example, can be seen at the level of the author’s ideas, emotions and experiences. Or at the level of the physical activities involved; muscle movements, using a keyboard or making notes. If you felt so inclined you could even analyse novel-writing in terms of the rearrangement of subatomic particles. An important point about looking at a phenomenon at different levels of complexity is that each level of complexity is equally valid. Novel-writing is about the author’s ideas, emotions and experiences. At the same time, it’s also about muscle movement and using a keyboard, and about the configuration of subatomic particles. But not all levels of complexity are equally useful for all purposes.

The way we organize knowledge

Levels of complexity have an important role in the way we organize our knowledge. Organizing knowledge is important because it reduces the amount of effort we have to put into thinking – our cognitive load. Instead of having to think about thousands of individual examples, we group things into categories and label the categories; ‘tigers’, ‘furniture’, ‘criminals’, ‘unacceptable behaviour’. For some entities we use informal ‘folk’ classifications that are loosely structured and often involve differences of opinion, such as the different types of criminal behaviour or which activities are acceptable and which are not. Sometimes there is general agreement on how to classify things and we develop formal, consistent classification systems – like the periodic table or the Linnaean taxonomy of biological organisms.

Carl Linnaeus’ classification of biological organisms has been in use for nearly 300 years. It’s stood the test of time because Linnaeus based it on the morphology (form) of organisms, rather than on their colour, habitat or uses, criteria that some previous classification systems had used. The form of organisms is very stable compared to their colour, habitat or what uses people make of them, which makes Linnaeus’ system highly reliable. His system has several levels of complexity called ‘ranks’, such as species, genus, family and class. The lowest rank is the species – organisms that are essentially identical in form. The higher the rank the more organisms are included in it. A genus might include only two species, whereas a class (e.g. mammals) might include thousands.

Levels of complexity are also referred to as levels of abstraction. That’s because each level has key defining features that can be abstracted out. The class ‘mammals’ for example, is defined by features such as the ability to regulate body temperature internally, having hair, mammary glands and a distinctive inner ear structure. On the face of it, abstracting key features appears to simplify matters because it tells you only about what all the members of the class have in common. But the class Mammalia includes thousands of species that possess many variations of the key features and lots of other features that aren’t common to all mammals, so although the shared features might be simple, a class could encompass a huge number of individual differences. A vet, called to treat an exotic species of squirrel, would find it useful to know that a squirrel is a mammal, but information about the squirrel’s species and about the individual squirrel’s medical history would be even more helpful.

Classifying behaviour

What has all this got to do with autism? Classification is important to autism because anything can be classified; chemical elements, biological organisms or human behaviour. Autism is defined in terms of behaviour – or more accurately, behavioural impairments. As far as I’m aware, although some people have devised classifications of behaviour for specific purposes, no one has produced the equivalent of a behavioural periodic table or Linnaean taxonomy. That’s because chemical elements and biological organisms are relatively stable things but behaviour isn’t. Although we can identify common behavioural patterns (eating, walking) and behavioural traits (neuroticism, aggression), behaviour tends to be very variable. Because of its variability there’s often disagreement about what constitutes a particular behaviour, so devising a formal, universally accepted classification system is problematic.

But behaviours can be classified, and at different levels of complexity, too. Take ‘feeding’ for example. Feeding is a label we attach to a set of behaviours that includes suckling, eating, drinking and foraging. ‘Eating’ can be broken down into more simple, lower level behaviours; opening the mouth, putting in food, masticating and swallowing. Each of these behaviours could be broken down even further if necessary, into the way specific muscles function, for example, or right down to the level of molecules or the ubiquitous subatomic particles. Which level of complexity we use to describe ‘eating’ will depend on why we are referring to that particular behaviour; a restaurant manager and physiotherapist will be interested in different levels.

The behaviours that are impaired in autistic disorder are social interaction, communication and behavioural flexibility – people diagnosed with autism show ‘restricted, repetitive and stereotyped patterns of behavior interests and activities’. There’s no doubt that those are features common to a large group of people. But each of those areas is a very large class of behaviours – it’s at a high level of complexity. This means that although two people might have totally different behavioural impairments, they can both qualify for a diagnosis of autism. It’s those individual differences I want to look at next.

gold image: Rob Lavinsky, – CC-BY-SA-3.0